Backed by 56,000+ GitHub stars and over 240 million Docker pulls, LiteLLM delivers the open-source AI gateway trusted by Netflix, Lemonade, Rocket Money, and thousands of engineering teams to route every LLM request through one unified API. The Rust-core gateway adds sub-millisecond overhead per request with 8ms P95 latency at 1,000 RPS, 15x throughput improvement and 11x lower memory footprint compared to Python-only proxies. A single OpenAI-compatible endpoint connects to 100+ providers and 1,800+ models spanning OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Vertex AI, Hugging Face, vLLM, Nvidia NIM, Ollama, and Mistral with day-zero support for new model releases. The Auto Router V2 classifies request complexity across four tiers using rule-based scoring, semantic keyword matching, and adaptive Thompson sampling to route each request to the most cost-effective model without API calls or training data. Virtual API keys enable multi-tenant governance with per-team, per-user, and per-project cost tracking, budget caps with automatic fallback rerouting, and role-based access control. Built-in guardrails provide PII masking, prompt injection detection, and model-graded evaluation before requests reach providers. The Agent Gateway extends routing from model calls to agent workflows with MCP server integration. Observability integrates with Langfuse, Arize Phoenix, OpenTelemetry, and MLflow for complete request tracing. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Ory Hydra is an OpenID Certified OAuth 2.0 and OpenID Connect provider that plugs into any existing user database or authentication system through a headless consent and login bridge, letting you own your authorization infrastructure without rebuilding identity management from scratch. The Go binary ships at under 15MB with zero system dependencies and exposes two distinct API surfaces: a public endpoint on port 4444 handling authorization code flows, token exchanges, PKCE challenges, device authorization grants per RFC 8628, and OpenID Connect discovery, plus an admin endpoint on port 4445 managing client registration, token introspection, consent session lifecycle, and JWKS rotation. Access tokens can be issued as opaque reference tokens or signed JWTs for stateless validation at resource servers, while refresh token rotation, token revocation per RFC 7009, and pairwise subject identifiers provide granular security controls. Persistent storage spans PostgreSQL, MySQL, and CockroachDB with automatic schema migrations, and the architecture supports horizontal scaling behind load balancers with shared database state. Dynamic client registration per RFC 7591 enables automated provisioning, and the OAuth 2.0 Threat Model security considerations are implemented as default behaviors rather than optional add-ons. Trusted by OpenAI and other internet-scale platforms. Helm charts and Docker Compose quickstarts provide production-ready deployment paths alongside the broader Ory ecosystem including Kratos for identity management, Oathkeeper for zero-trust API proxying, and Keto for fine-grained permissions. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Portabase takes a zero-trust approach to database backups: lightweight Rust/Tokio agents deploy next to each database, encrypt dumps with AES-GCM before data ever leaves the host, and poll the Next.js control plane outbound every five seconds requiring zero inbound firewall rules. This architecture contains blast radius if the dashboard is compromised while supporting ten engines with stable backup and restore: PostgreSQL 12 through 18, MySQL 5.7 through 9, MariaDB 10 and 11, MongoDB 4 through 8, SQLite 3.x, Redis 2.8+, Valkey 7.2+, Firebird 3.0 through 5.0, Microsoft SQL Server 2017 through 2022 including Azure SQL, and Docker volumes on Engine 20.10+. Encrypted backups store on configurable backends including local filesystems, any S3-compatible provider (AWS, MinIO, RustFS), Google Cloud Storage, and Azure Blob Storage. Cron-based scheduling with Grandfather-Father-Son retention policies automates backup lifecycle management, while on-demand restore targets any compatible server for cross-environment recovery. The CLI installs agents with a single command and auto-updates when new versions ship. A RESTful API with MCP server integration enables automation from CI/CD pipelines and AI agent workflows. Developed by a non-profit under Apache-2.0. 1,270+ stars and 174 releases since October 2024. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console.

With nearly 30,000 GitHub stars and deployment across organizations including Criteo, Pandora, and Barclays, HashiCorp Consul is the industry-standard platform for service discovery, service mesh, and distributed configuration across dynamic multi-cloud and multi-datacenter infrastructure. Services register themselves and become discoverable via a built-in DNS interface on port 8600 or an HTTP API on port 8500, with health checks ensuring only healthy instances receive traffic through automatic catalog deregistration and service-level circuit breaking. The service mesh capabilities use Envoy sidecar proxies with Transparent Proxy mode to establish automatic mTLS encryption for all service-to-service communication, while identity-based intentions define fine-grained authorization rules controlling which services can communicate. The integrated API Gateway manages north-south traffic into the mesh with configurable routing rules, TLS termination, and header-based matching policies. Consul's distributed key-value store provides hierarchical configuration storage accessible via CLI, HTTP API, and the built-in web UI, with blocking queries enabling watch-based configuration updates without polling. Multi-datacenter federation connects Consul clusters across regions through WAN gossip and RPC forwarding, enabling cross-datacenter service discovery and failover with configurable prepared queries. The Raft consensus protocol provides strong consistency for the service catalog and KV store, with anti-entropy mechanisms ensuring agent state converges with the server catalog. Consul integrates natively with Kubernetes via Helm charts with automatic sidecar injection, Nomad for workload orchestration, Vault for secrets management, and Terraform for infrastructure provisioning. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BUSL 1.1 licensed.

Powering notification delivery for thousands of organizations with over 39,000 GitHub stars, Novu is the open-source communication infrastructure that eliminates the need to build separate integrations for every messaging channel. The TypeScript-based platform provides a single API endpoint that routes notifications across In-App Inbox, Email via SendGrid, Mailgun, AWS SES, and Postmark, SMS through Twilio, Vonage, and Plivo, Push via Firebase Cloud Messaging, Expo, and APNS, and Chat through Slack, Discord, Microsoft Teams, and WhatsApp. The embeddable React Inbox component delivers real-time WebSocket-powered notifications with read/unread states, action buttons, and user-controlled preferences directly inside your application. The workflow engine orchestrates complex multi-channel delivery with conditional branching, time delays, digest batching to reduce notification fatigue, throttle controls, and per-step channel fallbacks. Novu Connect introduces Agent Communication Infrastructure enabling AI agents built with LangChain, Claude, or custom frameworks to hold threaded two-way conversations with humans across any supported channel through a single conversation model with identity resolution and credential management. The backend runs on Node.js with NestJS, MongoDB for notification storage, Redis with BullMQ for job queuing, and Socket.io for real-time delivery. Self-hosting deploys via Docker Compose with the dashboard accessible at port 4200. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Arcane gives you a single polished dashboard to manage Docker containers, images, volumes, networks, and Compose projects across unlimited remote hosts. The SvelteKit frontend paired with a Go backend delivers real-time resource monitoring with historical graphs, container lifecycle controls including shell exec and live log streaming, and one-click Docker Compose deployment with Git repository synchronization for version-controlled stack definitions. The manager-agent architecture connects remote environments via Direct TCP on port 3553 or Edge mode where agents initiate outbound gRPC/WebSocket connections through NAT and firewalls without requiring inbound ports, all secured with mTLS certificates. Vulnerability scanning identifies security issues in running container images directly from the interface. The backup system enables scheduled container snapshots with configurable retention for disaster recovery. Network and volume administration includes visual relationship mapping between services, and the responsive interface supports dark/light themes with full mobile optimization and community-driven internationalization via Crowdin. 6,500+ stars and 89 releases since April 2025 reflect a rapid development cadence. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSD-3-Clause licensed.

A status page and uptime monitor in one Go binary: Statping-ng - the actively maintained fork of Statping - replaces the UptimeRobot-plus-Statuspage combo with a ~20 MB Docker image using under 50 MB of RAM. It checks services over HTTP, TCP, UDP, ICMP ping, and gRPC health checks on configurable intervals, with per-service timeouts, expected status codes, POST requests with custom JSON bodies, SSL verification, and failure thresholds before alerting. The public status page is the differentiator against plain monitors: visitors see live status, uptime percentages, and latency charts grouped into service categories, with incident announcements and scheduled-maintenance messages you publish from the dashboard - and Sass-based custom styling matches the page to your brand rather than a vendor template. When something fails, notifiers fire immediately: Slack, Discord, Telegram, SMTP email, PagerDuty, Twilio SMS, Pushover, and custom webhooks, each testable before saving. Because notifiers are single Go files, the plugin system makes new channels straightforward. A RESTful API manages services and reads uptime data programmatically, and the free Statping mobile app connects to your server via QR code for on-the-go monitoring. Data persists to SQLite, MySQL, or PostgreSQL. Point it at internal services too - anything the container can reach is monitorable.

Powering Facebook, Uber, Twitter, and Booking.com, MySQL processes billions of queries daily as the world's most widely deployed open-source relational database. Its default InnoDB storage engine delivers full ACID compliance with row-level locking, crash recovery, and foreign key enforcement, making it the backbone of mission-critical e-commerce, SaaS, and financial applications. MySQL's Document Store bridges relational and NoSQL paradigms, allowing developers to work with schema-less JSON collections alongside traditional SQL tables through the X DevAPI protocol. For high availability, InnoDB Cluster combines Group Replication, MySQL Router, and MySQL Shell into an integrated solution with automatic failover, while InnoDB ClusterSet extends this across data centers for disaster recovery. The database supports multiple storage engines including MyISAM for read-heavy analytics, Memory for temporary high-speed caching, and Archive for compressed historical data. Native JSON functions like JSON_TABLE() transform document data into relational result sets, and spatial data types enable GIS applications. MySQL integrates with every major programming language through official connectors for Python, Java, Node.js, PHP, C++, and .NET, and works seamlessly with frameworks like Django, Laravel, Rails, and Spring. Replication topologies support read replicas for horizontal scaling, and the new Change Stream Applier in MySQL 26.7 enables configurable parallel transaction application with up to 1,024 workers per channel. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPLv2 licensed.

With over 430 GitHub stars and designed for DevOps engineers, platform teams, and homelab operators, Sencho is the self-hosted Docker Compose control plane that gives you a real operational cockpit without abandoning file-on-disk compose workflows. The web dashboard provides live CPU, memory, and disk sparklines per container, one-click start/stop/restart controls, and a Monaco-powered YAML editor with syntax highlighting, inline diff, and one-click rollback. Compose files remain the source of truth on the host filesystem. Multi-node management connects remote Sencho instances via authenticated HTTP and WebSocket proxy using long-lived API tokens, with no SSH access or exposed Docker sockets required. The Pilot Agent establishes an outbound-only WebSocket tunnel for nodes behind NAT, CGNAT, or strict firewalls. Blueprints define compose intent once and deploy it across label-targeted nodes, with drift detection keeping the fleet aligned. Fleet Federation provides cordon and pin controls for maintenance windows, while Fleet Actions execute bulk deploy, stop, and restart operations by stack label across the entire infrastructure. The security suite includes Trivy vulnerability scanning, deploy enforcement policies, SARIF and SBOM export, and Fleet Sync to replicate scan policies across replicas. Additional capabilities include 199+ one-click app templates, Git source integration, atomic deploys with auto-rollback, auto-heal, scheduled operations, webhooks, and custom S3 off-site backups to MinIO, R2, B2, or AWS. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

ToolHive is an open-source MCP server management platform that lets you run, secure, and orchestrate Model Context Protocol servers in isolated containers, giving AI agents structured access to tools like GitHub, databases, and cloud services without exposing host credentials or network. Every server launches inside its own sandboxed container with a minimal permission file, network access filtering, and encrypted secrets management, preventing misbehaving connectors from reaching beyond their defined scope. The built-in registry provides a catalog of vetted servers you can install with one command, while custom images and package-manager references let you onboard proprietary connectors without writing Dockerfiles. Platform teams deploy the Kubernetes operator to declare MCP servers as cluster resources using Custom Resource Definitions, with automated lifecycle management and multi-namespace isolation. The Virtual MCP Server gateway aggregates multiple backends behind a single endpoint, centralizing OIDC authentication, tool filtering, and composite cross-server workflows so clients connect once instead of juggling separate URLs. An MCP Optimizer analyzes tool schemas via semantic search and surfaces only relevant tools per request, cutting token consumption by up to 85%. OpenTelemetry traces and Prometheus metrics deliver full visibility into tool execution, latency, and request audit trails across every managed server. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Achieving 98.4% Recall@5 on LongMemEval-S without a reranker, AKB is the agent-native knowledge infrastructure giving Claude Code, Cursor, Windsurf, and any MCP-aware agent persistent organizational memory — a drop-in replacement for Confluence or Notion where AI agents read and write directly rather than scraping human-formatted pages. Documents are stored as Markdown with YAML frontmatter in Git bare repositories (full version history, branching, diff), while PostgreSQL 16 serves as source of truth for chunk text, metadata, and BM25 vocabulary, with dense embeddings handled by a pluggable vector store (pgvector default, Qdrant optional, or managed Seahorse). Agents interact through 20+ MCP tools — akb_put, akb_search, akb_browse, akb_relations, akb_sql, akb_edit, akb_history — performing hybrid dense plus BM25 search fused via Reciprocal Rank Fusion in one call. The URI graph (depends_on, related_to, implements relations) lets agents traverse knowledge connections without external graph databases. Multi-tenant vault isolation is enforced at the PostgreSQL ACL level with per-user roles and SET LOCAL ROLE for SQL sandboxing. Event fanout streams structured events to Redis Streams for external consumers (synthesis bots, digest agents, audit trails), and a hash-chained append-only audit log integrates with enterprise SIEMs. Ships with agent plugins for Claude Code and Codex (session capture, wiki ingest, lifecycle hooks), Kubernetes kustomize manifests, and Open Knowledge Format export. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSL 1.1 licensed (free under 100 seats, auto-converts to Apache 2.0 after four years).

Ralph tracks thousands of assets across data centers and back offices through a unified CMDB, DCIM, and asset management platform, providing the same capabilities as ServiceNow, Device42, and Sunbird dcTrack without per-asset licensing fees. Used by Allegro in production and backed by over 2,500 GitHub stars, it is built on Python 3 with Django and Django REST Framework. Interactive data center floor plans support drag-and-drop rack placement, front and rear rack views showing U-level asset positioning, power consumption tracking per PDU, and ethernet and fiber connection mapping. The back office module manages laptops, desktops, printers, mobile phones, and accessories with user assignment tracking, warehouse inventory, stock-taking audits, and automated loan workflows. IP address management handles network environments with DHCP and DNS integration alongside network topology visualization. Software license tracking monitors per-user and per-core allocations with usage calculations, compliance reporting, and expiration alerts, while vendor support contracts track SLA agreements and warranty renewal dates. The transitions system automates asset lifecycle processes through configurable workflows that trigger HTTP webhooks, custom Python scripts, email notifications, and Puppet and Ansible configuration management integration. Custom fields extend the data model with text, numbers, dates, choices, and URLs per asset type, accessible via the REST API. Multi-tenant architecture organizes assets by regions, data centers, service environments, and business units. Deploy via Docker with the allegro/ralph image alongside MySQL and Redis. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Vigil monitors your entire distributed infrastructure and generates a public status page from a single Rust binary small enough to run on a Raspberry Pi, consuming minimal CPU and memory while providing crash-free reliability. Four distinct monitoring modes cover every topology: poll probes check HTTP, TCP, SSH, and ICMP endpoints for reachability with configurable intervals and thresholds; push probes receive health reports from Vigil Reporter libraries embedded in your application code across Node.js, Python, Golang, Rust, TypeScript, Dart, and C#; local probes delegate monitoring to Vigil Local slave daemons running behind firewalls on separate LANs; and script probes execute custom shell commands for specialized health checks. Each monitored service transitions through healthy, sick, and dead states based on consecutive probe failures, with configurable thresholds controlling state transition sensitivity. When services change state, Vigil dispatches notifications through twelve alert channels including Slack, Email, Twilio SMS, Telegram, Pushover, Gotify, XMPP, Matrix, Zulip, Cisco Webex, and generic webhooks. The generated status page displays service groups organized by category with real-time replica status, system load metrics from reporter probes, and a maintenance announcement system for communicating planned downtime through the Manager HTTP API. Configuration uses a single TOML file defining all probes, services, and notification channels with no database dependency. Docker deployment pulls the official image with volume-mounted configuration. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MPL-2.0 licensed.

Tell an AI agent to deploy your app and watch it actually happen. Nixopus connects to your GitHub repo, analyzes the codebase to detect your framework, generates the right Docker configuration, resolves dependency conflicts, and ships the whole thing to your server with HTTPS enabled. When something breaks, the agent reads the logs, creates a pull request with a fix, and redeploys without waking you up. The chat interface lets you deploy, add custom domains, check logs, roll back, and troubleshoot in plain English from the web dashboard or directly from your VS Code or Cursor editor. Extensions bring one-click installs for 100+ self-hostable apps like Appwrite, Excalidraw, and Ollama, each showing you exactly what it runs on your server before installation. Connect multiple servers and monitor CPU, memory, disk, and running containers across your entire fleet from a single dashboard with draggable widget layouts. The built-in terminal opens SSH sessions right in the browser with split-pane support, so you can watch logs in one pane while running commands in another. Framework detection covers Next.js, Django, Rails, Go, FastAPI, and anything else that fits a container image. Every deployment gets automatic TLS via Caddy with Let's Encrypt certificates provisioned and renewed silently. Previous container images are retained so rollbacks skip the rebuild step entirely. Deploy on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Raw zone files and API calls become something a whole team can operate safely once PowerDNS-Admin puts its web interface in front of a PowerDNS authoritative server. It's a Python/Flask application covering full forward and reverse zone management, with the touches that matter in daily DNS work: zone templates for stamping out consistent new domains, easy IPv6 PTR record editing (reverse zones by hand are misery), full IDN/Punycode support for internationalized domains, and DynDNS 2 protocol support so routers and scripts can update records the way they would against a commercial dynamic-DNS service. Access control is enterprise-grade: local users, LDAP against OpenLDAP or Active Directory, SAML, and OAuth via Google, GitHub, Azure, or OpenID Connect, hardened with TOTP two-factor authentication. Role-based permissions extend to zone-specific access control - hand a developer their project's zone without exposing the rest of your namespace - and activity logging records who changed which record when, the audit trail bare PowerDNS never gives you. The dashboard monitors PDNS service configuration and statistics, and its own API exposes zone and record management for automation on top of the UI. Runs against MySQL/MariaDB or PostgreSQL, talking to PowerDNS through its REST API. MIT-licensed.

CoreObs delivers a unified self-hosted dashboard that replaces the typical combination of Uptime Kuma, Homer, and separate monitoring tools with a single interface for managing your entire server infrastructure. The Next.js frontend with shadcn components provides a modern dark-themed UI displaying server hardware metrics collected by a lightweight Go agent that leverages Glances for hardware abstraction — streaming real-time per-core CPU load, RAM utilization, NVIDIA GPU statistics, disk usage, system uptime, and load averages via WebSocket connections. The application registry tracks all self-hosted services with configurable uptime monitoring, availability history charts, and instant notifications through Discord, Telegram, Pushover, and email when services go down or recover. Quick-access links provide one-click navigation to each application's management panel directly from the dashboard. The network visualization module built on React Flow enables creating visual topology maps of your infrastructure with drag-and-drop nodes representing servers, switches, and services connected by labeled edges. Server data can be organized with tags, copied between entries, and monitored with configurable pagination and compact view modes. Deploy via Docker Compose with three containers — the Next.js web interface, the Go monitoring agent, and PostgreSQL 17 for persistent storage. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Born at Brazilian media giant Globo.com, Thumbor answers imaging CDNs like Imgix and Cloudinary with an HTTP service where every image variant is just a URL. Ask for /300x200/smart/your-image.jpg and Thumbor fetches the original, crops and resizes on demand, and caches the result - one source file, unlimited renditions, no batch pre-generation pipeline. The "smart" in the URL is the signature feature: OpenCV-based face detection finds people in the frame and crops around them (no more thumbnails with severed heads), and when no faces exist, feature detection finds visually important corners and computes a weighted center of mass as the focal point. Beyond cropping, a chainable filter pipeline handles brightness, contrast, grayscale, blur, red-eye removal, rounded corners, rotation, watermarks, and format conversion with quality control - applied in order via URL segments. All common image formats work out of the box, and every layer is pluggable: loaders (HTTP, local, S3), storages and result storages (local, S3, Ceph, and community backends), engines, optimizers, filters, and even custom detectors, with the awesome-thumbor list cataloging the ecosystem. URL signing prevents abuse of your processing capacity. Integrations exist for Django, Rails, Node, WordPress, and most frameworks. MIT-licensed, battle- tested for over a decade.

ServerBee delivers complete fleet monitoring from a single Rust binary that embeds both the server and web UI, requiring no external database or heavy runtime. Lightweight agents installed on each node stream CPU, memory, disk, network, load, temperature, GPU, and disk I/O metrics over persistent WebSocket connections to a central server backed by embedded SQLite. The real-time React dashboard presents 17 drag-and-drop widget types across custom layouts, with historical charts spanning one hour to 30 days and monthly traffic statistics with billing-cycle prediction and cost insights including burn rate and per-resource unit cost scoring. Network monitoring covers ICMP, TCP, and HTTP ping probes, service monitors for SSL certificates, DNS records, HTTP keyword matching, TCP ports, and WHOIS expiration, plus network-quality probing across 96 China 3-ISP and international presets with IP-quality and streaming-unlock detection. Remote management features include a web terminal with sandboxed file manager, Docker container stats with logs and events, firewall configuration, and agent auto-update capabilities. Public status pages display 90-day uptime timelines, while alerting supports email and webhook notifications with flexible trigger expressions. A native iOS companion app provides mobile monitoring access. Deploy via Docker or download the single binary for Linux, macOS, or Windows, running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.